北京生物医学工程

最优阈值生长和形态学结合的肺气道树分割方法

Pulmonary Airway Tree Segmentation by Combining Optimal Threshold Region Grow and Morphology

作者：王昌黄煜峰王兴家冯焕清李传富

单位：中国科学技术大学电子科学与技术系(合肥230027)

关键词：灰度尺度重建；肺部气道树分割；最优阈值区域生长；高分辨率CT；形态学算子

分类号：

出版年·卷·期（页码）：2010·29·3（241-244）

摘要：

在肺气道树分割的过程中,由于部分容积效应和噪声污染的影响,容易出现支气管断裂和分割泄漏现象，因此不能分割出精确肺部气道树。为此本文提出一种最优阈值生长和形态学结合的气道树分割方法。首先利用最优阈值生长算法分割初略的肺部气道树，利用灰度重建的形态学算子提取潜在的精细肺气管区域，然后将上述两种分割结果合成一个完整的肺部气道树，最后利用种子点区域生长法去除结果中的伪气管区域，得到包含第5级以及约60%第6级的支气管。本方法有效解决了高精度肺气道树分割中的支气管断裂和泄漏问题，有较好的鲁棒性。

In the process of pulmonary airway tree segmentation, it is prone to bronchial rupture and segmentation leakage, during to partial volume effects and noise pollution. In this paper, we proposed the hybrid method which was composed of optimal threshold region grow and morphology for pulmonary airway tree segmentation. Firstly the optimal threshold region growth algorithm was used to obtain low-level of airway tree, and the grayscale reconstruction morphological operator was used to extract fine potential regions of airway. Then by combining these two results of segmentation method, we obtained the integrated pulmonary airway tree. Lastly the method of region grow was used on integrated data set to remove pseudo-tracheal regions to ensure the three-dimensional connectivity of airway tree, including the trachea at level 5 and about 60% of the trachea at level 6. The proposed method effectively solved the problems of bronchial rupture and segmentation leakage, in the segmentation of high-precision airway tree, and had a good robustness.

参考文献：

［1］Juerg Tschirrena, Hoffman EA. Airway tree segmentation using adaptive regions of interest Proceedings of SPIE, 2004, 5369:117-124.
［2］Atilla PK, William E, Higgins GM. Three-dimensional Human Airway Segmentation Methods for Clinical Virtual Bronchoscopy. Academic Radiology, 2002, 9(10):1153-1167.
［3］赵鹏, 冯焕清, 李传富,等. 一种基于参数自适应的肺气道树分割方法. 北京生物医学工程,2009,28(5):454-457.
［4］Schlathter T,Lorenz C,Carlsen S,et al. Simultaneous segmentation and tree reconstruction of the airways for virtual bronchoscopy. SPIE Medical Imaging, 2002, 4684:103-113.
［5］Cohen LD, Deschamps T. Segmentation of 3D tubular objects with adaptive front propagation and minimal tree extraction for 3D medical imaging. Computer Methods in Biomechanics and Biomedical Engineering, 2007. DOI: 10.1080/10255840701328239.
［6］Chandrasekhar P,Zerhcuni E, Mitzner W.Segmentation of 3d pulmonary trees using mathematical morphology. CHANDRASEKHAR PISUPATI ET AL http://citeseerx.ist.psu.edu/view.
［7］Fetita, Preteux CI. Pulmonary Airways: 3-D Reconstruction From Multislice CT and Clinical Investigation. IEEE Trans on Medical Imaging, 2004, 23(11):1353-1363
［8］Preteux F. On a distance function approach for gray-level mathematical morphology. //Edward Dougherty, ed.Mathematical morphology in image processing.New York: M. Dekker, 1993.
［9］Francoise P, Capderou A, Grenier P. Modeling, segmentation, and caliber estimation of bronchi in high resolution computerized tomography. Journal of Electronic Imaging, 1999, 8(1):35-46.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录